Experience and Educationis the best concise statement on education ever published by John Dewey, the man acknowledged to be the pre-eminent educational theorist of the twentieth century. Written more than two decades after Democracy and Education(Dewey's most comprehensive statement of his position in educational philosophy), this book demonstrates how Dewey reformulated his ideas as a result of his intervening experience with the progressive schools and in the light of the criticisms his theories had received. Analysing both "traditional" and "progressive" education, Dr. Dewey here insists that neither the old nor the new education is adequate and that each is miseducative because neither of them applies the principles of a carefully developed philosophy of experience. Many pages of this volume illustrate Dr. Dewey's ideas for a philosophy of experience and its relation to education. He particularly urges that all teachers and educators looking for a new movement in education should think in terms of the deeped and larger issues of education rather than in terms of some divisive "ism" about education, even such an "ism" as "progressivism." His philosophy, here expressed in its most essential, most readable form, predicates an American educational system that respects all sources of experience, on that offers a true learning situation that is both historical and social, both orderly and dynamic.

经验与教育 = Experience and education

“Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks”の学習報告

Matrix Factorization Techniques for Recommender Systems

Extending beyond the boundaries of science, art, and culture, content-based multimedia information retrieval provides new paradigms and methods for searching through the myriad variety of media all over the world. This survey reviews 100p recent articles on content-based multimedia information retrieval and discusses their role in current research directions which include browsing and search paradigms, user studies, affective computing, learning, semantic queries, new features and media types, high performance indexing, and evaluation techniques. Based on the current state of the art, we discuss the major challenges for the future.

/pdf/content-based-multimedia-information-retrieval-state-of-the-41z647ll8w.pdf

Content-based multimedia information retrieval: State of the art and challenges

A grammar of spoken Chinese = 中國話的文法

It has been very difficult to develop spoken dialogue systems with high domain extensibility. Not only the system complexity inevitably increases with the number of topics and domains, but the concurrent topics need to be handled persistently and consistently across different domains. This paper presents a distributed architecture for cooperative spoken dialogue agents with high domain extensibility. Under this architecture, different spoken dialogue agents handling different domains can be developed independently, and cooperate with one another to respond to the user’s requests, while a user interface agent can access the correct spoken dialogue agent through a domain switching protocol, and carry over the dialogue state and history so as to keep the knowledge processed persistently and consistently across different domains. Initial experiments provide very encouraging results.

A distributed architecture for cooperative spoken dialogue agents with coherent dialogue state and history

This paper presents an entropy-based algorithm for accurate and robust endpoint detection for speech recognition under noisy environments. Instead of using the conventional energy-based features, the spectral entropy is developed to identify the speech segments accurately. Experimental results show that this algorithm outperforms the energy-based algorithms in both detection accuracy and recognition performance under noisy environments, with an average error rate reduction of more than 16%.

/pdf/robust-entropy-based-endpoint-detection-for-speech-19dgg90lt0.pdf

Robust entropy-based endpoint detection for speech recognition in noisy environments.

Three classes of subsyllabic units for Mandarin syllables are defined, i.e., the initials, the finals, and the transitions. A new structure for Mandarin syllable recognition is developed, in which the tones and base syllables are recognized jointly and a total of 574 subsyllabic unit models will be enough to provide improved recognition performance. This approach has the potential to be extended to continuous Mandarin speech recognition, although the preliminary experiments demonstrated here are based on isolated syllables only. >

http://ntur.lib.ntu.edu.tw/bitstream/246246/2007041908562353/1/00319276.pdf

A new framework for recognition of Mandarin syllables with tones using sub-syllabic units

A successfully implemented real-time Mandarin dictation machine which recognizes Mandarin speech with unlimited texts and very large vocabulary for the input of Chinese characters to computers is described. Isolated syllables including the tones are first recognized using specially trained hidden Markov models with special feature parameters. The exact characters are then identified from the syllables using a Markov Chinese language model. The real-time implementation is on an IBM PC/AT, connected to a set of special hardware boards on which ten TMS 320C25 chips operate in parallel. It takes only 0.45 s to dictate a character. >

/pdf/a-real-time-mandarin-dictation-machine-for-chinese-language-5arz0duz98.pdf

A real-time Mandarin dictation machine for Chinese language with unlimited texts and very large vocabulary

The vector representations of fixed dimensionality for words (in text) offered by Word2Vec have been shown to be very useful in many application scenarios, in particular due to the semantic information they carry. This paper proposes a parallel version, the Audio Word2Vec. It offers the vector representations of fixed dimensionality for variable-length audio segments. These vector representations are shown to describe the sequential phonetic structures of the audio segments to a good degree, with very attractive real world applications such as query-by-example Spoken Term Detection (STD). In this STD application, the proposed approach significantly outperformed the conventional Dynamic Time Warping (DTW) based approaches at significantly lower computation requirements. We propose unsupervised learning of Audio Word2Vec from audio data without human annotation using Sequence-to-sequence Audoencoder (SA). SA consists of two RNNs equipped with Long Short-Term Memory (LSTM) units: the first RNN (encoder) maps the input audio sequence into a vector representation of fixed dimensionality, and the second RNN (decoder) maps the representation back to the input audio sequence. The two RNNs are jointly trained by minimizing the reconstruction error. Denoising Sequence-to-sequence Autoencoder (DSA) is furthered proposed offering more robust learning.

Lin-Shan Lee

Papers

A distributed architecture for cooperative spoken dialogue agents with coherent dialogue state and history

Robust entropy-based endpoint detection for speech recognition in noisy environments.

A new framework for recognition of Mandarin syllables with tones using sub-syllabic units

A real-time Mandarin dictation machine for Chinese language with unlimited texts and very large vocabulary

Audio Word2Vec: Unsupervised Learning of Audio Segment Representations Using Sequence-to-Sequence Autoencoder.